Identifying Opinion Holders and Targets with Dependency Parser in Chinese News Texts

Abstract

In this paper, we propose to identify opinion holders and targets with dependency parser in Chinese news texts, i.e. to identify opinion holders by means of reporting verbs and to identify opinion targets by considering both opinion holders and opinion-bearing words. The experiments with NTCIR-7 MOAT’s Chinese test data show that our approach provides better performance than the baselines and most systems reported at NTCIR-7. 1